[ogg-dev] eos on continued page

[Ralph Giles]

On 25-Feb-08, at 9:58 PM, Conrad Parker wrote:

> I'd definitely prefer that the recommendation to not have an
> incomplete packet on a page marked eos is a SHOULD not a MUST, so that
> eos pages don't need to be completely rebuilt after chopping.

Yes. The general guideline has been not to prohibit anything  
unnecessarily, but also to allow possible variations to communicate  
something. E.g. a stream SHOULD have an eos page, but if it doesn't  
it's an indication that the stream (still valid) has been truncated,  
and it's reasonable for a parser to report that information. I'd  
treat packet data after eos similarly.

  -r